Machine learning for classifying tuberculosis drug-resistance from DNA sequencing data.
Machine learning for classifying tuberculosis drug-resistance from DNA sequencing data.
复制标题
DOI：
10.1093/bioinformatics/btx801
复制
发表时间：
2018-05-15
期刊：
影响因子：
--
通讯作者：
Clifton DA
中科院分区：
文献类型：
--
作者：
Yang Y;Niehaus KE;Walker TM;Iqbal Z;Walker AS;Wilson DJ;Peto TEA;Crook DW;Smith EG;Zhu T;Clifton DA
正确、快速地测定结核分枝杆菌(MTB)对现有结核病药物的耐药性对于结核病的控制和管理至关重要。传统的分子诊断试验假设任何研究良好的单核苷酸多态的存在都足以引起耐药性，这就导致了对耐药性分类的低敏感性。考虑到结核分枝杆菌DNA测序数据的可用性，我们建立了1839株英国细菌的机器学习模型，以分类结核分枝杆菌对8种抗结核药物(异烟肼、利福平、乙胺丁醇、吡津酰胺、环丙沙星、莫西沙星、氧氟沙星、链霉素)的耐药性，并对多重耐药性进行分类。与以前基于规则的方法相比，表现最好的模型对异烟肼、利福平和乙胺丁醇的敏感度分别增加了2-4%，达到97%(P<；0.01)；对于环丙沙星和多药耐药结核病，敏感度增加到96%。对莫西沙星和氧氟沙星的敏感性分别从83%和81%增加到95%和96%(P<；0.01)。特别是，我们的模型与以前基于规则的方法相比，对吡津酰胺和链霉素的敏感度分别提高了15%和24%，提高了84%和87%(P<；0.01)。表现最好的模型使吡嗪酰胺和链霉素的ROC曲线下面积增加了10%(P<；0.01)，其他药物的ROC曲线增加了4-8%(P<；0.01)。有关源代码的详细信息，请访问http://www.robots.ox.ac.uk/~davidc/code.php.补充数据可在生物信息学在线上获得。
Correct and rapid determination of Mycobacterium tuberculosis (MTB) resistance against available tuberculosis (TB) drugs is essential for the control and management of TB. Conventional molecular diagnostic test assumes that the presence of any well-studied single nucleotide polymorphisms is sufficient to cause resistance, which yields low sensitivity for resistance classification. Given the availability of DNA sequencing data from MTB, we developed machine learning models for a cohort of 1839 UK bacterial isolates to classify MTB resistance against eight anti-TB drugs (isoniazid, rifampicin, ethambutol, pyrazinamide, ciprofloxacin, moxifloxacin, ofloxacin, streptomycin) and to classify multi-drug resistance. Compared to previous rules-based approach, the sensitivities from the best-performing models increased by 2-4% for isoniazid, rifampicin and ethambutol to 97% (P < 0.01), respectively; for ciprofloxacin and multi-drug resistant TB, they increased to 96%. For moxifloxacin and ofloxacin, sensitivities increased by 12 and 15% from 83 and 81% based on existing known resistance alleles to 95% and 96% (P < 0.01), respectively. Particularly, our models improved sensitivities compared to the previous rules-based approach by 15 and 24% to 84 and 87% for pyrazinamide and streptomycin (P < 0.01), respectively. The best-performing models increase the area-under-the-ROC curve by 10% for pyrazinamide and streptomycin (P < 0.01), and 4–8% for other drugs (P < 0.01). The details of source code are provided at http://www.robots.ox.ac.uk/~davidc/code.php. Supplementary data are available at Bioinformatics online.
登录
查看更多内容
DOI：
10.1073/pnas.1611283113
发表时间：
2016-11-29
影响因子：
11.1
作者：
Eldholm, Vegard;Pettersson, John H. -O.;Balloux, Francois
通讯作者：
Balloux, Francois
对来自中国的 161 株结核分枝杆菌分离株进行基因组测序，鉴定出与耐药性相关的基因和基因间区域
影响因子：
30.8
作者：
Zhang, Hongtai;Li, Dongfang;Bi, Lijun
通讯作者：
Bi, Lijun
影响因子：
30.8
作者：
Comas, Inaki;Borrell, Sonia;Roetzer, Andreas;Rose, Graham;Malla, Bijaya;Kato-Maeda, Midori;Galagan, James;Niemann, Stefan;Gagneux, Sebastien
通讯作者：
Gagneux, Sebastien
影响因子：
3.7
作者：
Stucki D;Malla B;Hostettler S;Huna T;Feldmann J;Yeboah-Manu D;Borrell S;Fenner L;Comas I;Coscollà M;Gagneux S
通讯作者：
Gagneux S
影响因子：
3.7
作者：
Georghiou SB;Magana M;Garfein RS;Catanzaro DG;Catanzaro A;Rodwell TC
通讯作者：
Rodwell TC